🧵
I've taught @CUBoulder since 2001 but I really didn't think much about diversity until I starting teaching sports governance @CUBuffs Athletics in 2014
I moved from teaching almost exclusively white kids to classes that actually looked like America
That woke me up
Some data...
At @CUBoulder Arts & Sciences (my campus home) we had 16,235 full-time undergrad students in Fall 2019
Of those, 463 (3.0%) were African American
% from 2018 to 2019 dropped ~6%
My program @CUBoulderENVS (Environmental Studies) is not diverse
We have 792 full-time undergrads
7 (0.9%) are African-American
None in our freshmen class
At @CUBoulder we have 3,538 total "academic" staff
Of those, 46 (1.3%) are African American
In 2009 it was 1.4%
These data are just embarrassing & wrong
The university Covid response risks making them worse (not just on my campus)
We have got to do better
Going from ENVS-->Athletics-->ENVS helped wake me up
We need to do better & I am committed to it
/END
🧵
The percentage of a percentage trick is increasingly common & leads to massive confusion
Here a undetectable difference of 0.01 events per year per decade is presented as the difference between a 31% and 66.4% increase (in the *likelihood* of the event, not the event itself)
The resulting confusion is perfectly predictable
Here is a reporter (NPR) explaining completely incorrectly:
"The phenomenon has grown up to 66% since the mid-20th century"
False
Also, the numbers in the text and figure do not appear to match up
I asked Swain about this over at BlooSkeye
A Frankenstein dataset results from splicing together two time series found online
Below is an example for US hurricane damage 1900-2017
Data for 1980-2017 was replaced with a different time series in the green box
Upwards trend results (red ---)
Claim: Due to climate change!
The errors here are so obvious and consequential that it is baffling that the community does not quickly correct course
The IPCC AR6 cited a paper misusing the Frankenstein hurricane loss dataset to suggest that NOAA's gold standard hurricane "best track" dataset may be flawed
JFC - Using flawed economic loss data to suggest that direct measurements of hurricanes are in error!
We’ve reached the point where an IPCC author is openly rejecting the conclusions of the IPCC out of concern over how their political opposition is correctly interpreting the AR6
The integrity of the IPCC on extreme events is now under attack
The IPCC explains that a trend in a particular variable is DETECTED if it is outside internal variability and judged with >90% likelihood
For most (not all) metrics of extreme weather detection has not been achieved
That’s not me saying that, but IPCC AR6
The IPCC also assesses that for most (but not all) metrics of extreme weather the signal of a change in climate will not emerge from internal variability with high confidence (ie, >90%) by 2050 or 2100, even assuming the most extreme changes under RCP8.5
The US National Academy of Sciences has a new study committee on Extreme Event Attribution
Among its sponsors are the Bezos Earth Fund and Robert Litterman
Who are they? . . .
The Bezos Earth Fund sponsors World Weather Attribution, an advocacy group promoting the connection of weather events w/ fossil fuels in support of press coverage & lawsuits
Robert Litterman is on the board of Climate Central which founded WWA & collaborates on climate advocacy
The fact that a NAS committee is funded by political advocates is crazy enough
But that is not all
On the committee itself are individuals from two climate advocacy groups
One . . . the Union of Concerned Scientists which is working to use attribution to support lawsuits . . .